ViewTube

Python 3.14 makes it easy to invoke a version of the interpreter without the Global Interpreter Lock (GIL). This makes it especially ...

Eight cores. The same script, running twice. In the first run, you see one core at 100% and the other seven dormant. In the ...

Your multi-threaded Python code is likely performing 70% slower than it should because of a 30-year-old architectural ghost.

You paid for eight cores and Python runs on one. The reason is a single mutex called the Global Interpreter Lock — and it isn't a ...

What Is the Global Interpreter Lock (GIL) in Python? If you've ever wondered why your Python multithreaded programs don't run ...

8 cores, 8 threads — zero speedup. Meet the GIL: one lock that has held all of Python for 30 years. Why it was put there, why it ...

Stop guessing which concurrency model to use in Python. We break down the technical differences between Threading and ...

The Global Interpreter Lock has defined Python's threading limitations for decades — but not anymore. In this deep dive, we ...

... Topics Covered Python, Python GIL, Global Interpreter Lock, Python threading, Python multithreading, Python multiprocessing, ...

Making sense of concurrency in Python 3.14 - Michael Seifert - PyCon Italia 2026 Elevator Pitch: Python 3.14 gives us additional ...

Python Global Interpreter Lock (GIL) – Deep Dive In this video, we take a deep dive into Python's Global Interpreter Lock (GIL) ...

EuroPython 2026 - S1 on 2026-07-17] *Speeding Up Python with Free Threading and Mypyc by Jukka Lehtosalo* ...

This lesson explains why: the Global Interpreter Lock. You'll see why threads still help for I/O-bound work despite the GIL, why ...

Are we free-threaded ready? Looking at where free-threaded Python fails - Cheuk Ho - PyCon Italia 2026 Elevator Pitch: ...

... the Global Interpreter Lock (GIL) in mind: the event loop runs all of your async in a single thread, regardless of the GIL presence.

Python threads vs processes, benchmarked live: threads made CPU-bound code 3% faster, processes made it 3.5x faster; then for ...